Eplerenone and New-Onset Diabetes in Mild Heart Failure
Eplerenone and New-Onset Diabetes in Mild Heart Failure
That diabetes is common in patients with heart failure was confirmed in the present study. Thirty-two per cent of participants had diabetes at baseline, and in those without diabetes at baseline the incidence of new diabetes was 21 cases per 1000 patient-years of follow-up. This incidence is similar to that reported in two other clinical trials in chronic heart failure and is considerably higher than expected in the general population, e.g. the incidence of physician-diagnosed diabetes was 12.8 cases per 1000 patient-years in American adults aged 65–79 years in 2009.
The effect of MR antagonists on the risk of diabetes is difficult to predict. Primary hyperaldosteronism is a state of insulin resistance, and aldosterone levels are associated with insulin resistance in normotensive and hypertensive patients. In an observational study of patients with primary hyperaldosteronism, treatment with either spironolactone or adrenalectomy returned insulin sensitivity parameters towards normal. In another study, the combination of vitamin E and spironolactone for 8 weeks decreased insulin levels and Homeostasis Model Assessment of Insulin Resistance (HOMA-IR) compared with vitamin E alone in patients with biopsy-proven NAFLD, while spironolactone therapy has shown benefical cardiac effects in patients with metabolic syndrome. Low serum potassium is also known to be associated with increased risk of new-onset diabetes in both normotensive and treated hypertensive patients, and MR antagonists increase potassium. On the other hand, spironolactone therapy has consistently led to modest increases in HbA1c in clinical trials of patients with type 2 diabetes alone or complicated by nephropathy or poorly controlled hypertension. In a further study of 107 patients with mild chronic heart failure, spironolactone therapy increased HbA1c and cortisol levels and reduced adiponectin levels over 4 months, findings which might be expected to herald an increased risk of developing diabetes. In contrast, in that study eplerenone had no such effects, suggesting the possibility of a differential effect, depending on the selectivity of MR blockade. Consistent with this, we found no effect of eplerenone on incident diabetes in the present study, although the number of affected patients was relatively small.
We identified seven factors independently associated with new-onset diabetes. Of these, longer duration of heart failure, higher serum ALT, and higher waist circumference were expected predictors. Serum ALT has been associated with incident diabetes in a number of studies in several different populations, including heart failure. Usually this is thought to be because ALT is considered a surrogate marker for NAFLD, a condition characterized by insulin resistance and therefore a powerful risk factor for diabetes. Whether or not this is the explanation in heart failure is less certain because transaminases may be elevated due to liver congestion or ischaemia. Interestingly, waist circumference performed slightly better as a predictor compared with either BMI or body weight. Digitalis was strongly predictive of new-onset diabetes, a finding noted in earlier studies but unexplained. One possibility is that digitalis use is a marker of heart failure severity, and there is some evidence that the risk of diabetes increases with increasing severity of heart failure. However, it has also been suggested that reduced activity of Na/K-ATPase may play a role in the pathogenesis of diabetes mellitus, and digoxin is an inhibitor of this enzyme. Lower age was also associated with higher risk of developing diabetes. Though at first sight surprising, this was noted in previous studies and seems likely to reflect either the different phenotypes of older and younger trial participants [data exist to support this: in a substudy of 2107 patients in CHARM, the prevalence of obesity (BMI ≥30 kg/m) was four times higher in patients with chronic heart failure aged <50 years than in patients aged ≥80 years] or the longer survival of younger participants recruited in trials of chronic heart failure patients. When the independent variables were combined in predictive models, they provided reasonable c-statistics of 0.74–0.76. Indeed, this is comparable with models for prediction of diabetes in the general population, despite not containing any direct glucose-related measures.
Various strengths and weaknesses of the data contained in this report warrant discussion. This is the first randomized trial to report the effect of MR antagonist therapy on new-onset diabetes, and participants were well phenotyped at baseline. While diabetes was physician-diagnosed and not based on trial-required measurements of fasting plasma glucose, participants were specifically asked about new-onset diabetes at each trial visit. Data for family history of diabetes were not available. The numbers of participants that developed diabetes was relatively small, so statistical power to detect a modest effect of eplerenone on new-onset diabetes was limited.
In summary, information available from commonly measured clinical and laboratory variables predicts the development of new-onset diabetes in patients with symptomatically mild systolic heart failure with similar accuracy to models used in other populations. Treatment of these patients with eplerenone had no effect on the development of diabetes compared with placebo.
Discussion
That diabetes is common in patients with heart failure was confirmed in the present study. Thirty-two per cent of participants had diabetes at baseline, and in those without diabetes at baseline the incidence of new diabetes was 21 cases per 1000 patient-years of follow-up. This incidence is similar to that reported in two other clinical trials in chronic heart failure and is considerably higher than expected in the general population, e.g. the incidence of physician-diagnosed diabetes was 12.8 cases per 1000 patient-years in American adults aged 65–79 years in 2009.
The effect of MR antagonists on the risk of diabetes is difficult to predict. Primary hyperaldosteronism is a state of insulin resistance, and aldosterone levels are associated with insulin resistance in normotensive and hypertensive patients. In an observational study of patients with primary hyperaldosteronism, treatment with either spironolactone or adrenalectomy returned insulin sensitivity parameters towards normal. In another study, the combination of vitamin E and spironolactone for 8 weeks decreased insulin levels and Homeostasis Model Assessment of Insulin Resistance (HOMA-IR) compared with vitamin E alone in patients with biopsy-proven NAFLD, while spironolactone therapy has shown benefical cardiac effects in patients with metabolic syndrome. Low serum potassium is also known to be associated with increased risk of new-onset diabetes in both normotensive and treated hypertensive patients, and MR antagonists increase potassium. On the other hand, spironolactone therapy has consistently led to modest increases in HbA1c in clinical trials of patients with type 2 diabetes alone or complicated by nephropathy or poorly controlled hypertension. In a further study of 107 patients with mild chronic heart failure, spironolactone therapy increased HbA1c and cortisol levels and reduced adiponectin levels over 4 months, findings which might be expected to herald an increased risk of developing diabetes. In contrast, in that study eplerenone had no such effects, suggesting the possibility of a differential effect, depending on the selectivity of MR blockade. Consistent with this, we found no effect of eplerenone on incident diabetes in the present study, although the number of affected patients was relatively small.
We identified seven factors independently associated with new-onset diabetes. Of these, longer duration of heart failure, higher serum ALT, and higher waist circumference were expected predictors. Serum ALT has been associated with incident diabetes in a number of studies in several different populations, including heart failure. Usually this is thought to be because ALT is considered a surrogate marker for NAFLD, a condition characterized by insulin resistance and therefore a powerful risk factor for diabetes. Whether or not this is the explanation in heart failure is less certain because transaminases may be elevated due to liver congestion or ischaemia. Interestingly, waist circumference performed slightly better as a predictor compared with either BMI or body weight. Digitalis was strongly predictive of new-onset diabetes, a finding noted in earlier studies but unexplained. One possibility is that digitalis use is a marker of heart failure severity, and there is some evidence that the risk of diabetes increases with increasing severity of heart failure. However, it has also been suggested that reduced activity of Na/K-ATPase may play a role in the pathogenesis of diabetes mellitus, and digoxin is an inhibitor of this enzyme. Lower age was also associated with higher risk of developing diabetes. Though at first sight surprising, this was noted in previous studies and seems likely to reflect either the different phenotypes of older and younger trial participants [data exist to support this: in a substudy of 2107 patients in CHARM, the prevalence of obesity (BMI ≥30 kg/m) was four times higher in patients with chronic heart failure aged <50 years than in patients aged ≥80 years] or the longer survival of younger participants recruited in trials of chronic heart failure patients. When the independent variables were combined in predictive models, they provided reasonable c-statistics of 0.74–0.76. Indeed, this is comparable with models for prediction of diabetes in the general population, despite not containing any direct glucose-related measures.
Various strengths and weaknesses of the data contained in this report warrant discussion. This is the first randomized trial to report the effect of MR antagonist therapy on new-onset diabetes, and participants were well phenotyped at baseline. While diabetes was physician-diagnosed and not based on trial-required measurements of fasting plasma glucose, participants were specifically asked about new-onset diabetes at each trial visit. Data for family history of diabetes were not available. The numbers of participants that developed diabetes was relatively small, so statistical power to detect a modest effect of eplerenone on new-onset diabetes was limited.
In summary, information available from commonly measured clinical and laboratory variables predicts the development of new-onset diabetes in patients with symptomatically mild systolic heart failure with similar accuracy to models used in other populations. Treatment of these patients with eplerenone had no effect on the development of diabetes compared with placebo.